In Project 2, we have shifted our focus to genetic analyses of positional candidate genes encoding proteins implicated in presenilin-related pathways. Over the past decade, our group has led the ascertainment, evaluation, and comprehensive genetic analyses of the NIMH Genetics Initiative Alzheimer's disease (AD) sample. This set of AD families, which currently includes 1527 individuals in 457 families, is the largest uniformly ascertained and evaluated sample ever assembled for the study of AD genetics. In separately funded studies, our laboratory recently completed the genetic analyses and preliminary follow-up studies of a comprehensive high-resolution genome screen for novel AD genes. The genome screen revealed that in addition to the expected highly significant linkage peak on chromosome 19q (APOE locus), several regions demonstrate evidence of 'suggestive' linkage to AD. In separately funded studies, we are further refining these linkage regions using additional markers. The immense data emanating from this screen constitute a powerful resource for research efforts aimed at identifying novel AD genes. To identify novel AD genes, we propose a positional candidate approach in which we test biologically compelling candidate genes from our AD genetic linkage peaks using family-based association, augmented by linkage disequilibrium mapping. Clusters of tightly spaced single nucleotide polymorphisms (SNPs) in positional candidate genes encoding proteins that are biologically implicated in presenilin-related pathways will be tested for association with AD. These genes have been divided into six categories: a. presenilin interactors, b. PS/gamma-secretase substrates, c. PS/gamma-secretase complex components and proteins that affect PS/gamma-secretase activity/Abeta generation, d. amyloid precursor protein (APP) C-terminus/APP intracellular domain (AICD) interactors, e. AICD transcriptional targets, and f. presenilin-like signal peptide peptidases. Testing of these genes will be prioritized according to their genomic position relative to the best-confirmed AD genetic linkage peaks, and their biological relevance to AD pathogenesis. Candidate genes for which positive signals are successfully found in the NIMH sample will be tested in an independent sample, the Consortium on Alzheimer's Genetics (CAG) sample. SNPs found to be strongly associated with AD in these samples will be phenotypically and functionally characterized including collaborative analyses with the six other PPG laboratories.